Synthesising Oscillators

[Freywa]

(look at X66 for an example: 166 gliders to 13)

I wasn't aware there had been a 166-glider one. Do you have a reference to it? I'd be curious to see it, for historical reasons.

Even that information is a bit out of date; there's a 12-glider synthesis now.

Here's Extrementhusiast's original 166-glider synthesis -- except the post says it's 146 gliders. That's clearly not right, but I don't know whether 166 is the right number either -- I can only find 162 gliders in that initial incremental synthesis. Anyone want to build a continuous synthesis and make sure?

Why do you keep posting the same information I posted twice?

[hkoenig]

Because "what I tell you three times is true."

[Freywa]

I've extended my synthesis cache to the end of 2014 and the 7-glider twin bees shuttle by Martin, filling in some of the 2013 syntheses I had missed. The files are now sorted by year, since a handful of them are from years after 2014.

I also found a few inconsistencies in LifeWiki's synthesis data:

• Bob Shemyakin came up with a 6-glider synthesis of tumbler before Aidan F. Pierce did.
• 21P2's cost is listed as 7. This is clearly a copy-paste error, as the RLE file shows that 7 gliders is only for the 1 beacon; an eighth glider turns this into 21P2.

The files ending with "_synth" are ready to be included in LifeWiki. 88P28 and pentoad 2 have all the necessary comment lines, but are still not ready because I have not figured out the exact glider cost; the RLEs announcing them only showed the final step.

Some of the missing 2013 syntheses I pinpointed are the less common p2 oscillators from Flammenkamp's census. Because they are natural objects, there's a good chance that I may have missed later improvements to these objects, in which case you should let me know.

P.S. Now up to 25 June 2015 with 9-glider synthesis of caaba (originally 10, but the initial constellation is in 3), a name I just concocted out of its apgcode:

Code: Select all

#N caaba_synth
#O Martin Grant, 25 June 2015
#O Jeremy Tan, 19 March 2019
#C 9-glider synthesis of caaba
#C http://conwaylife.com/wiki/Xp2_caabaiczw32
#C http://conwaylife.com/forums/viewtopic.php?p=20721#p20721
#C http://conwaylife.com/forums/viewtopic.php?p=73418#p73418
x = 110, y = 43, rule = B3/S23
65bo$65bobo$bo63b2o$2bo$3o79bo$82bobo$82b2o3$78b3o2$76bo5bo$29bo8bo37b
o5bo$30b2o4b2o38bo5bo$29b2o6b2o$78b3o$31bo18bo$31b2o7bo8bobo$30bobo7bo
8bobo$40bo9bo2$16bo19b3o3b3o$17b2o72b2o13b2o$16b2o22bo50bobo12bobo$40b
o47b2obobo9b2obobo$40bo47bo2bob2o8bo2bob2o$15b2o74bobo12bobo$16b2o72bo
2bo11bo2bo$15bo75b2o13b2o12$62b2o$62bobo$62bo!

[Freywa]

At last! At long last! I've run all the way through this thread and collated all the notable oscillator syntheses and improvements!

…now I'm dead tired and my eyes hurt.

[dvgrn]

I've run all the way through this thread and collated all the notable oscillator syntheses and improvements...

Awesome! And super quadruple thanks for the additional comments, and for adding a _synth suffix and standard comments to most of the files. I have a multi-hour train ride this morning, so I'll do a quick review of all of these now, but it looks like I can just upload most of them without any editing.

EDIT: Do you happen to have a list of the glider count for each of these syntheses? I'm just going to upload all the files for now, but eventually someone will have to go through and add

Code: Select all

|synthesis        = N
|synthesisRLE     = true

to every article, or edit N if there's an existing synthesis. For the files I've looked at so far it doesn't look like that detail is included in the comments.

[Bullet51]

I've run all the way through this thread and collated all the notable oscillator syntheses and improvements!

Wow. What a great compilation!
By the way, the Pentoad II synthesis could be largely simplified:

Code: Select all

x = 12, y = 19, rule = B3/S23
7bo$5bobo$5bo2bo$5b3o$10bo$9bobo$9bobo$10bo4$bo$obo$obo$bo$4b3o$3bo2bo
$4bobo$4bo!

or maybe

Code: Select all

x = 12, y = 27, rule = B3/S23
2bobo$4bo$3obo2$6bo$5b3o$5b3o3$9b3o$9b3o6$3o$3o3$4b3o$4b3o$5bo2$7bob3o
$7bo$7bobo!

Code: Select all

x = 10, y = 17, rule = B3/S23
8b2o$8b2o4$3o4b2o$3o4bobo$obo$2o6b2o$7bobo$obo4b3o$b2o4b3o4$2o$2o!

Code: Select all

x = 17, y = 12, rule = B3/S23
6bo$2o3bobo$2o2bo$5bo2bo$6b3o2b2o$11b2o$4b2o$4b2o2b3o$8bo2bo$12bo2b2o$
9bobo3b2o$10bo!

[Freywa]

By the way, the Pentoad II synthesis could be largely simplified:

Code: Select all

x = 12, y = 27, rule = B3/S23
2bobo$4bo$3obo2$6bo$5b3o$5b3o3$9b3o$9b3o6$3o$3o3$4b3o$4b3o$5bo2$7bob3o
$7bo$7bobo!

You gave me exactly the right idea. Pentoad 2 in 16 gliders:

Code: Select all

x = 68, y = 37, rule = B3/S23
28bo$21bobo2bobo$22b2o3b2o$22bo10bobo$33b2o$9bo24bo$7b2o$8b2o$25b2o5bo
$6bo17bobo6bo2b3o$7bo18bo4b3o23b2o$5b3o49b2o2$39bo18bo$7b2o29bobo16b2o
$7bobo28bobo15bobo$7bo31bo16b2o$63bo$60b4o$60bo$2bo27bo35b2o$obo26bobo
33bobo$b2o26bobo33b2o$30bo34bo2$2b3o60b2o$2bo33b3o4bo21b2o$3bo27b3o2bo
6bobo$37bo5b2o$2o$b2o$o34bo$35b2o$34bobo10bo$41b2o3b2o$41bobo2bobo$41b
o!

[Freywa]

EDIT: Do you happen to have a list of the glider count for each of these syntheses? I'm just going to upload all the files for now, but eventually someone will have to go through and add

Code: Select all

|synthesis        = N
|synthesisRLE     = true

to every article, or edit N if there's an existing synthesis. For the files I've looked at so far it doesn't look like that detail is included in the comments.

I included a count of gliders as the first #C line in each RLE:

Code: Select all

#N blocksongriddle_synth
#O Martin Grant, 26 August 2015
#C 5-glider synthesis of blocks on griddle
#C http://conwaylife.com/wiki/Blocks_on_griddle
#C http://conwaylife.com/forums/viewtopic.php?p=22172#p22172
x = 29, y = 24, rule = B3/S23
7bobo15bo$8b2o15bobo$8bo14bo4bo$23b6o2$23b2o2b2o$9bo13b2o2b2o$7bobo2bo
$8b2o2bobo$12b2o4$b2o$obo$2bo6$14b2o$14bobo$14bo!

Since you want all of those counts in one convenient location, I wrote a small script (gc.py) which prints out the counts and has been placed in the attachment to this post, which now has glider counts for every one of its patterns (and has been updated with my 16-glider pentoad 2 synthesis, the last pattern for which I added the count). The counts are reported in the file called out, whose contents I reproduce below:

Code: Select all

106p135: 29
124p37: 76
132p37: 40
204p47: 142
22p2: 5
23p2: 10
24p10: 15
24p2: 15
26p2: 41
26p40: 10
28p6: 14
28p73: 30
34p13: 14
38p111: 136
38p72: 14
44p123: 79
44p72: 127
48p31: 24
53p13: 219
56p27: 20
56p29: 43
60p312: 24
60p33: 16
78p70: 35
88p25: 36
88p28: 92
92p156: 55
blocksongriddle: 5
burloaferimeter: 23
caaba: 9
catereron36p22: 20
cisbeacononanvil: 22
cyclic: 10
diuresis: 20
eaterblockfrob: 8
eureka: 12
fox: 7
harbor: 23
hectic: 26
hertzoscillator: 64
hexapole: 9
hooks: 15
jasonsp36caterers: 28
koksgalaxy: 11
lightweightemulator: 111
loaflipflop: 20
oddtesttubebaby: 7
p35honeyfarmhassler: 19
p40bheptominoshuttle: 29
p56bheptominoshuttle: 15
p60bheptominohassler: 28
p60trafficlighthassler: 69
p60trafficlighthasslerpd: 30
pennylane: 27
pentant: 29
pentoad1h2: 19
pentoad2: 16
period156glidergun: 16
prepulsarshuttle47: 52
quadpole: 5
roseluck: 29
shortkeys: 6
skewedquad: 8
thumb1: 98
thunderbird: 3
tnosedp4: 56
tnosedp6: 32
tricetongs: 10
tumbler: 6
twinbeesshuttle: 7
twirlingttetsons2: 40
twotransgriddleswithtwotubs: 9

The keys refer to the file names, which may differ from the actual pattern names. For example. 34p13 refers to 34p13.rle, which is about Beluchenko's p13.

[Ian07]

I went through the wiki and added most of these syntheses to the infoboxes (dvgrn still needs to actually upload them). Just a question though: does it matter if the synthesis produces a different stator variant of the oscillator? Asking this due to 204P47, 88P25, Diuresis, Hertz oscillator, and Jason's p36, in which the pattern synthesized in the file is different from the one in the infobox. There's also p60trafficlighthasslerpd_synth.rle, which currently isn't on the wiki at all AFAIK.

[dvgrn]

I went through the wiki and added most of these syntheses to the infoboxes (dvgrn still needs to actually upload them). Just a question though: does it matter if the synthesis produces a different stator variant of the oscillator? Asking this due to 204P47, 88P25, Diuresis, Hertz oscillator, and Jason's p36, in which the pattern synthesized in the file is different from the one in the infobox. There's also p60trafficlighthasslerpd.rle, which currently isn't on the wiki at all AFAIK.

I ran into the problem with 204p47_synth.rle, but in that case the variant in question is shown under Pre-pulsar shuttle 47. So in that case I figured it was fine to upload the _synth file, but it wouldn't change the cost parameter for the basic shuttle.

It should be okay to have {pname}_synth files without a matching article with {pname} in the infobox; we can just add a synthesis link in the image or LifeViewer caption. Maybe mention the cost as part of that link?

EDIT: On another topic, should we change the "38p111" pname to "38p11.1"? A discussion a while back concluded that it would more or less okay to include periods in pnames, just for the few cases where they would avoid confusion due to two numbers getting concatenated into larger numbers. At least, a period seems to me like the lesser of two evils, since otherwise the pname here seems to promising a period-111 oscillator.

[Ian07]

I ran into the problem with 204p47_synth.rle, but in that case the variant in question is shown under Pre-pulsar shuttle 47. So in that case I figured it was fine to upload the _synth file, but it wouldn't change the cost parameter for the basic shuttle.

Hmm, I noticed that the pname for the synthesis doesn't match the pname for the pattern itself, which is actually prepulsarshuttle47v2.rle.

It should be okay to have {pname}_synth files without a matching article with {pname} in the infobox; we can just add a synthesis link in the image or LifeViewer caption. Maybe mention the cost as part of that link?

Sounds good. I'm thinking we can format it inside the caption like this:

Code: Select all

'''142-[[glider]] synthesis''' {{JavaRLE|204p47_synth|brief}}

EDIT: On another topic, should we change the "38p111" pname to "38p11.1"? A discussion a while back concluded that it would more or less okay to include periods in pnames, just for the few cases where they would avoid confusion due to two numbers getting concatenated into larger numbers. At least, a period seems to me like the lesser of two evils, since otherwise the pname here seems to promising a period-111 oscillator.

Also a very good idea in my opinion. There are of course probably several more pnames to be changed for this to be consistent across the whole collection.

[dvgrn]

Hmm, I noticed that the pname for the synthesis doesn't match the pname for the pattern itself, which is actually prepulsarshuttle47v2.rle.

Re-uploaded under the name prepulsarshuttle47v2_synth.rle. A similar thing happened with the pname twoblockershasslingrpentomino: the template system wouldn't have known what to do with roseluck_synth.rle

I also got the rest of the files uploaded, I think. If anything looks wrong that I should repair, please let me know. Now my eyes are too tired to stare at this stuff any more tonight.

[Freywa]

I'm just going to leave this here:

Code: Select all

x = 494, y = 328, rule = B3/S23
123bo$124bo141bo$122b3o142bo$265b3o2$52b2o48b2o149b2o149bo12bo38b2o$
52bobo48bo2b2o144bo2bo147bobo12bo36bo2bo21bobo$54bo48bobo2bo17bo126b3o
146bo2bo10b3o2bo33bob2o21b2o$54bobo12bobo32bo2b2o18b2o148bo124b3o16b2o
30b2o2bo23bo$55bobo12b2o54b2o127b3o15bo2bo143bobo3bobo23bo4bo20bo$56bo
13bo49b2o8bobo122bo2bo12bobo2b3o125b3o19b2o24bob3o19bobo$121b2o7b2o
124b2o14b2o129bo2bo20bo2b3o20b2o22b2o$120bo10bo271bobo24bo$404bo26bo
55b3o$69b2o4bo411bo$70b2o2bo191b2o204b3o13bo$69bo4b3o190b2o205bo$266bo
206bo2$74b2o$74bobo$74bo10$469bo$467bobo$468b2o5$469bo2bobo$278bo46bo
141bobo2b2o$252b2o3bo20bobo22bo19b2o28b2o102bo10b2o3bo$252bobobobo19b
2o22bobo19b2o26bo2bob2o19bo4bo68b2o2bobo$254bobobo15bo27bo2bob2o19b2o
23bobobobo19b2o2bobo66bo3bo2bo$253b2ob2o17bo27bobobo19b2o25b2obobo18b
2o3b2o68b3ob2o$273b3o28bo2bo21bo28bo96bo$305b2o16b2o131b3o$324b2o65bo
66bo$323bo66b2o$269b2o3bo53b2o60bobo$268bobo2b2o52b2o143b2o$270bo2bobo
53bo143b2o$472bo2$375bo$375b2o$374bobo11$17bobo$18b2o$18bo2$468bo$174b
o45bo200bobo42bobo$174bobo41bobo3bo197b2o43b2o$25bobo2bo138bobo2b2o43b
2o4b2o46bo148bo$26b2o2bobo137b2o52b2o45bobo204bo$26bo3b2o120b2ob2o13bo
31b2o48b2o18b2o2bobo123b2o48b2o3b2o15bo2bo$152b2obo27bo19bo2bo21bobo
21bobo21b2o124bobob2o17bo26bobobo2bo12bobo2b3o$155bo26b2o19bobobo15b3o
2b2o24bo22bo126bobobo14b2o28b2obob2o13b2o$2o153bobo24bobo17b2obo2bo14b
o5bo23b2ob2o13b2o129bobobobo15b2o30bobo$obo153b2o48b2o16bo31bobo13b2o
128b2o3bo20b2o26bobo$3bobo160b2o88bobo12bo5b3o142bo5bobo26bo18b3o$4b2o
159bobo89bo19bo144b2o4bo49bo$167bo110bo142bobo53bo$24b3o$26bo$25bo18$
70bobo$71b2o$71bo$81bobo$81b2o141bobo$82bo142b2o2bobo$225bo3b2o$52b2o
149b2o25bo$52bobo2b2o143bobo20bo$55bo2bo6b2o3bo131bobob2o16b2o$56b2o6b
obo2b2o132b2obobo15bobo$66bo2bobo135bo11bo$219b2o$218bobo23$27bo$25b2o
$26b2o2$23bo$18b3o3b2o$2bob2o14bo2b2o377b2o3bo$2b2o2bo12bo382bobobobo$
5bobo396bobobo11bobo$6bo14b3o380bo2bo13b2o$21bo380bobo16bo$22bo379b2o$
423b2o$424b2o$423bo3$417bo14b2o$417b2o12b2o$416bobo14bo12$271bo$225bo
46bo$225bobo42b3o$225b2o149bo$377bo$273bobo7bo91b3o$224b3o4b3o39b2o7b
2o$218b2o6bo4bo42bo7bobo$217bobo5bo6bo187bobo$219bo201b2o$202b2o3b2o
44b2o16b3o79bo17b2o3bo28bo15bo$202bobobobo45bo2b2o14bo78bobo17b2o2bobo
26b3o17bo$204bobo45bobobobo13bo80bobo15bo4b2o25b2o3bo14b2o$204bobo45b
2o2bo98bo46bo2bobobo14b2o6bo$205bo50b2o97bob2o43bo2bob2o22b2o$353bobob
2o12b2o30b2o26bobo$353b2o15bobo48bo$372bo48b2o$420bobo18$73bobo$74b2o$
74bo2$223bobo201bo$5bo217b2o202bobo39bo$3bobo218bo202b2o41bo$4b2o70bob
o2bo386b3o$77b2o2bobo41bobo287bo2bobo$8bo43b2o23bo3b2o20b2o21b2o74b2o
149b2o19bobo2bo22b2o3bo5bobo2b2o33b2o$6b2o19b2o23bobo47bo2bo20bo75bo2b
2o145bo2bo19b2o2bobo20bobobobo5b2o3bo32bo2bo$7b2o17bo2bo23bobo46bobobo
15bo80bobobo22bo121b2obo19bo3b2o23bobo2bo43b2o$2b2o22bobo25bobo46bo2bo
15b2o4bobo71b2obobo21b2o124b2o47bo2b2o$3b2o20b2ob2o25bobo46b2o15bobo4b
2o8bo67bo18b2o2bobo120b2obo46bobo48b4o$2bo53bo72bo7b2o85bobo125bo2bo
46b2o48bo4bo16bo$11b2o124bobo86bo126b2o72b2o24b4o15bobo$11bobo413bobo
43b2o2bo$11bo63b3o155b2o137bo6b3o45bo27b2o20bobo$77bo154b2o138b2o5bo
75b2o20b2o$76bo157bo136bobo6bo2$470bo$470b2o$469bobo18$479bo$477bobo
11bo$478b2o11bobo$454b2ob2o32b2o$453bobob2o24b2o$453bobo28b2o$452b2obo
b4o22bo$456b2o2bo$486bo$485b2o$485bobo21$374bo$375b2o$374b2o4$372b2o$
352b2o17bobo2bobo$352bobo18bo2b2o$354bo22bo$352bo2b2o$352b2obo$355bo$
355bobo$356b2o13b2o$372b2o$371bo16$377bobo$378b2o$378bo4$379bo$379bobo
$379b2o$387bo$353b2o31b2o$352bo2bo26b2o2bobo$352bo2bo25bobo$353b2o28bo
2$353b4o$353bo2bo$354b2o!

These are the first batch of 4-glider syntheses of Bob Shemyakin, filtered out to exclude those objects that appear in Hickerson's 2,3,4-glider synthesis stamp collection. I have a view to adding these syntheses to the collection as well where they improve on known syntheses or where the synthesis is not on LifeWiki.

[Freywa]

So I eventually decided to comb through the 4-glider thread and add any still lifes that got synthesised in 4 gliders. And then I found the soup search results thread, added a few more syntheses from there and corrected a few errors (including my idiosyncratic names and the periods).

There's just so many new and improved syntheses out there that haven't been catalogued in the LifeWiki that I can't possibly find them all alone. But the attachment here ought to be a start. (For still lifes, Catagolue has syntheses for many small objects, but not necessarily the most efficient one – as I found out when I discovered the long integral synthesis in 4 gliders from simeks.)

Code: Select all

106p135: 29
124p37: 76
12p2: 8
132p37: 40
14p2.1: 8
18p2.471: 9
22p2: 5
23p2: 10
24p10: 15
24p2: 15
26p2: 41
26p40: 10
28p6: 14
28p7.3: 30
34p13: 14
38p11.1: 136
38p7.2: 14
44p12.3: 79
44p7.2: 127
48p31: 24
53p13: 219
56p27: 20
56p29: 43
60p312: 24
60p33: 16
78p70: 35
88p25: 36
88p28: 92
92p156: 55
beehiveoncap: 4
beehiveondock: 4
bicap: 4
bipond: 3
blocksongriddle: 5
burloaferimeter: 23
catereron36p22: 20
cisbeacononanvil: 22
cishookandrbee: 4
cishookwithtail: 5
cisrotatedhook: 4
cyclic: 10
diuresis: 20
eaterblockfrob: 8
eureka: 12
fox: 7
harbor: 23
hectic: 26
hertzoscillator: 64
hexapole: 8
honeycomb: 9
hooks: 15
jasonsp36caterers: 28
koksgalaxy: 11
krake: 4
lightweightemulator: 111
loaflipflop: 20
long3barge: 4
long3boat: 4
long3ship: 5
longclawwithtail: 4
longintegral: 4
longshillelagh: 4
loop: 4
oddtesttubebaby: 7
p35honeyfarmhassler: 19
p40bheptominoshuttle: 29
p56bheptominoshuttle: 15
p60bheptominohassler: 28
p60trafficlighthassler: 69
p60trafficlighthasslerpd: 30
pennylane: 27
pentant: 29
pentapole: 9
pentoad1h2: 19
pentoad2: 16
period156glidergun: 16
prepulsarshuttle47: 52
prepulsarshuttle47v2: 142
quadpole: 5
shortkeys: 6
skewedquad: 8
symmetricscorpion: 4
thirteenloop: 4
thumb1: 98
thunderbird: 3
tnosedp4: 56
tnosedp6: 32
transrotatedrbee: 4
tricetongs: 10
tubwithlongtail: 4
tumbler: 6
twelveloop: 4
twinbeesshuttle: 7
twirlingttetsons2: 40
twoblockershasslingrpentomino: 29
twotransgriddleswithtwotubs: 9
verylongshillelagh: 4

I tried but failed to find efficient syntheses for the following:

• Two cis griddles with two tubs
• 14P2.3 and 14P2.4

[Goldtiger997]

Nice work on doing this!

I tried but failed to find efficient syntheses for the following:

• Two cis griddles with two tubs
• 14P2.3 and 14P2.4

You can find syntheses for those in the collection of small oscillator syntheses (up to 16 bits) that I compiled.

Two cis griddles with two tubs:

Code: Select all

x = 108, y = 26, rule = B3/S23
21b2o20bo17bo11bo29bo$20b2o21bobo16bo10bobo27bobo$18bo3bo18bo4bo13b3o
8bo4bo24bo4bo$16bobo23b4obo17b2o5b4obo22bob4obo$17b2o22bo4bo17b2o5bo4b
o24bo4bo$43bobo20bo6bobo27bobo$43bo17b3o9bo29bo$16b3o44bo$18bo43bo$17b
o3$3o19bo$2bo18b2o$bo19bobo$15bo$15b2o$14bobo6$42b2o$41b2o$43bo!

14P2.3:

Code: Select all

x = 116, y = 20, rule = B3/S23
34bo$35b2o17bo$29bo4b2o18bobo25bo$27bobo24b2o24bobo$28b2o5bo45b2o$35b
2o$34bobo26bo19bo$63bo19bo$63bo19bo4$35b2o36b3o17b3o17b3o$obo31bobo32b
2o18b2o18b2o$b2o19bo13bo5bo26bobob2o14bobob2o14bobob2o$bo19bobo17bobo$
22b2o18b2o27bobo17bobo17bobo$b3o66bo19bo19bo$bo68b2o18b2o18b2o$2bo!

14P2.4:

Code: Select all

x = 421, y = 54, rule = B3/S23
357bo$358b2o$357b2o2$384bo7bobo$355bo27bo8b2o4bo$353bobo27b3o7bo4bobo$
354b2o42b2o8$188bo$187bo$129bo57b3o45bo$125bo2bo56bo47bobo57bo$65bobo
58bob3o55bo47b2o55bobo$66b2o56b3o57b3o28b2o28b2o28b2o15b2o11b2o28b2o
38b2o38bo$66bo148b2o18bo9b2o18b2o8b2o18b2o8b2o28b2o38b2o38bo$72bo24b2o
28b2o106b2o27bo2bo26bo2bo28b2o38b2o45bo2bo$9bo61bo25bobo27bobo28b2o28b
2o28b2o14bobo11b2o14bo2bo10b2o14bo2bo10b2o15bobo10b2o25bobo10b2o38bo$
8bo62b3o24bobo27bobo27bobo27bobo27bobo27bobo14b2o11bobo14b2o11bobo14b
2o11bobo24b2o11bobo32b2o3bobo$8b3o64b3o21bobo27bobo27bobo27bobo27bobo
27bobo27bobo27bobo27bobo37bobo37bo$37bo29bo7bo21bobobo25bobobo25bobobo
25bobobo25bobobo25bobobo25bobobo25bobobo25bobobo35bobobo33bobobo$7bo
27b3o27b3o8bo18b3ob2o24b3ob2o24b3ob2o24b3ob2o24b3ob2o24b3ob2o24b3ob2o
24b3ob2o24b3ob2o34b3ob2o36bo$6b2o26bo29bo29bo29bo29bo29bo29bo29bo29bo
29bo29bo39bo$6bobo24bobo27bobo27bobo27bobo27bobo27bobo27bobo27bobo27bo
bo27bobo27bobo37bobo$2o31b2o28b2o28b2o28b2o28b2o28b2o28b2o28b2o28b2o
28b2o28b2o38b2o$b2o$o393b2o$393b2o$4b3o388bo$6bo$5bo353b2o$358bobo40b
2o$353bo6bo40bobo$353b2o46bo$352bobo$396b2o$365bo29b2o$366bo24b2o4bo$
364b3o24bobo$351b3o37bo$353bo3b2o$352bo5b2o11bo$357bo14b2o$371b2o2$
369b2o$368bobo$370bo!

(For still lifes, Catagolue has syntheses for many small objects, but not necessarily the most efficient one – as I found out when I discovered the long integral synthesis in 4 gliders from simeks.)

The reason for this is because chris_c's display_synth script (which is what catagolue uses) hasn't been updated since May 2017. However, it does not seem trivial to update, since chris_c has described losing a lot of the related scripts in a "backup blunder", so a lot of the scripts would have to be re-written.

Edit: I just noticed chris_c updated the readme only two months ago, which now exlpains how the min-paths.txt synthesis storage format works in detail.

[dvgrn]

So I eventually decided to comb through the 4-glider thread and add any still lifes that got synthesised in 4 gliders. And then I found the soup search results thread, added a few more syntheses from there and corrected a few errors (including my idiosyncratic names and the periods).

Thanks again! I've uploaded all the changes I could find.

The new additions are all mixed in with files already uploaded, with no easy list to follow. So I just uploaded the ones with a 3/20 date stamp: syntheses for twelveloop, hexapole, longintegral, pentapole, 18p2.471, twocisgriddleswtwotubs and 14p2.3 and 14p2.4 from Goldtiger997, and almost everything in the 2014 subfolder. If I've missed any, let me know.

[Freywa]

So I eventually decided to comb through the 4-glider thread and add any still lifes that got synthesised in 4 gliders. And then I found the soup search results thread, added a few more syntheses from there and corrected a few errors (including my idiosyncratic names and the periods).

Thanks again! I've uploaded all the changes I could find.

The new additions are all mixed in with files already uploaded, with no easy list to follow. So I just uploaded the ones with a 3/20 date stamp: syntheses for twelveloop, hexapole, longintegral, pentapole, 18p2.471, twocisgriddleswtwotubs and 14p2.3 and 14p2.4 from Goldtiger997, and almost everything in the 2014 subfolder. If I've missed any, let me know.

As it turns out, I left for dinner at my grandmother's house shortly after posting the updated syntheses collection, during which time you made your comment. I still had one more file to add after accounting for the "missing" syntheses kindly provided by Goldtiger997 – that for blinkers bit pole.

Maintaining the compilation here has become unwieldy, so I have created a folder in my Dounreay git repository on GitLab to hold them. This also allows for pull requests should you come up with a new or improved synthesis not found on the LifeWiki. The blinkers bit pole file has been included in that repository folder as of this post.

It should be relatively easy to determine which files have been changed, and therefore should be uploaded, by comparing the filenames between versions.

[dvgrn]

It should be relatively easy to determine which files have been changed, and therefore should be uploaded, by comparing the filenames between versions.

Thanks, a Git repository is definitely a huge improvement once it gets going. It will be easy to get a list of recently changed files just by checking the commit history.

I've uploaded all the new files including blinkersbitpole_synth.rle, and I updated the LifeWiki article for that one. I haven't yet edited the articles to point to the following syntheses, with the correct number of gliders:

... twelveloop, hexapole, longintegral, pentapole, 18p2.471, twocisgriddleswtwotubs and 14p2.3 and 14p2.4 from Goldtiger997, and almost everything in the 2014 subfolder (everything except short keys and tumbler)

If anyone wants to take over that job, I'd be most grateful -- I'm about out of time for today again.

[mniemiec]

11-glider synthesis of Muttering Moat 1, down from 21, inspired by the third soup https://catagolue.appspot.com/object/xp ... 3201/b3s23 (although all three seem to make it a similar way) It's remotely possible that there is a one-glider cleanup; I found one that eliminates 3 of the 4 still-lifes, but that doesn't help at all.

Code: Select all

x = 158, y = 27, rule = B3/S23
63bo$61boo$62boo$$55bo62bo$56boo60bobo$55boo27boo28boobboo$83bobbo26bo
bbo$84boo28boo$62bobo$62boo23bo29bo$63bo23bo29bo$56bobo28bo29bo$57boo$
57bo$5bo85boo28boo28boo$3bobo27boo28boo26boboboo14bobo7boboboo24bobob
oo$oobboo27boo28boo31bo15boo12bo29bo$boo4boo29boo28boo22bobbo16bo9bobb
o26bobbo$o5bobo29boo16bo11boo15bo7bo21bo7bo29bo$8bo47boo27bo7bobobo17b
o7bobobo25bobobo$55bobobboo23bo10boo17bo10boo28boo$59boo$61bo$$87boo
28boo$87boo28boo!

[Freywa]

11-glider synthesis of Muttering Moat 1, down from 21, inspired by the third soup https://catagolue.appspot.com/object/xp ... 3201/b3s23 (although all three seem to make it a similar way) It's remotely possible that there is a one-glider cleanup; I found one that eliminates 3 of the 4 still-lifes, but that doesn't help at all.

Code: Select all

x = 158, y = 27, rule = B3/S23
63bo$61boo$62boo$$55bo62bo$56boo60bobo$55boo27boo28boobboo$83bobbo26bo
bbo$84boo28boo$62bobo$62boo23bo29bo$63bo23bo29bo$56bobo28bo29bo$57boo$
57bo$5bo85boo28boo28boo$3bobo27boo28boo26boboboo14bobo7boboboo24bobob
oo$oobboo27boo28boo31bo15boo12bo29bo$boo4boo29boo28boo22bobbo16bo9bobb
o26bobbo$o5bobo29boo16bo11boo15bo7bo21bo7bo29bo$8bo47boo27bo7bobobo17b
o7bobobo25bobobo$55bobobboo23bo10boo17bo10boo28boo$59boo$61bo$$87boo
28boo$87boo28boo!

Thanks, I've updated the git folder.

Edit: Six Ls in 20 gliders, based on this soup:

Code: Select all

x = 153, y = 48, rule = B3/S23
32bo$33bo28bo16bobo$31b3o26bobo16b2o$61b2o17bo2$76bo$76bobo$76b2o3$61b
o70bo$62bo68bo$15bo44b3o16bo51b3o$14bo64bobo$14b3o46bo15b2o47b2o$62bob
o63b2o$11b3o27bobo18bobo$11bo30b2o19bo$12bo29bo2$118bo30bo$115bo2b3o
25bo2b3o$115b3o3bo24b3o3bo$119b3o28b3o$115b3o28b3o$115bo3b3o24bo3b3o$
116b3o2bo25b3o2bo$118bo30bo2$4bo73bo$5bo51bo19b2o$3b3o50bobo18bobo$56b
obo48b2o$3o37b2o15bo49b2o$2bo36bobo$bo39bo16b3o42b3o$58bo46bo$59bo44bo
3$43b2o$42bobo$44bo2$40bo17b2o$40b2o16bobo26b3o$39bobo16bo28bo$88bo!

You should be assuming that my syntheses sub-repository is being updated with each new or improved synthesis I find or discover.

Edit 2: Loading dock in 14 gliders, again soup-inspired (and with a little help from UnLife):

Code: Select all

x = 83, y = 30, rule = B3/S23
15bo$13b2o$10bo3b2o14bo15bo$bobo4bobo20bo15bo$2b2o5b2o18b3o13b3o$2bo
12bo$13b2o18b2o13bo$14b2o17b2o8b2o2bobo$43b2o2bobo$48bo2$47bo30bo$45b
3o28b3o$b2o41bo3b2o25bo3b2o$obo15bo24bob2o3bo23bob2o3bo$2bo15bobo23bo
3b2obo23bo3b2obo$18b2o25b2o3bo25b2o3bo$47b3o28b3o$47bo30bo2$46bo$45bob
o2b2o$5b2o38bobo2b2o8b2o$6b2o38bo13b2o$5bo12bo$10b2o5b2o28b3o12b3o$10b
obo4bobo27bo14bo$5b2o3bo37bo14bo$6b2o$5bo!

Edit 3: Snake dance in 16 gliders from this soup:

Code: Select all

x = 69, y = 47, rule = B3/S23
28bo$26b2o$27b2o5$17bo$17b2o$16bobo7bo$26bobo$26b2o5$26bo10bo$24bobo
11b2o$obo7bo14b2o10b2o$b2o8bo50bob2o$bo7b3o4b3o43b2obo$18bo46bob2o$17b
o42b3obo2bo$60bo2bobo2bo$29bo31bo2bob3o$28bo31b2obo$28b3o4b3o7bo17bob
2o$35bo8b2o17b2obo$8b2o10b2o14bo7bobo$7b2o11bobo$9bo10bo5$19b2o$18bobo
$20bo7bobo$28b2o$29bo5$18b2o$19b2o$18bo!

Edit 4: The original diamond ring in 24 gliders from this soup, beating Jason Summers's 28G for a stator variant:

Code: Select all

x = 206, y = 67, rule = B3/S23
88bo$86b2o$70bo16b2o$68bobo$69b2o$155bo$154bo$154b3o$150b2o$150b2o4$
156b2o$156b2o6$132bobo$71b3o59b2o$34bo21bo16bo59bo9bo$32b2o23bo14bo70b
o$33b2o20b3o85bo$37b2o46b2o$29bo6b2o16bo30bo53b3o27b2o$28b2o8bo15b2o
32bo61bo18b2o5bo22bo$28bobo22bobo31b2o60bobo24bobo19bobo$148bobobo23b
2o19bobobo$99bo48bo3bo44bo3bo$100bo45b2obobob2o40b2obobob2o$98b3o44bo
2bobobo2bo18b2o18bo2bobobo2bo$125b2o17bobo2bobo2bobo17b2o17bobo2bobo2b
obo$60b3o62b2o18bo2bobobo2bo38bo2bobobo2bo$60bo85b2obobob2o40b2obobob
2o$61bo86bo3bo44bo3bo$123b2o23bobobo44bobobo$8bobo61b2o31bobo14bobo24b
obo46bobo$o8b2o61bo32b2o17bo5b2o18bo48bo$b2o6bo65bo30bo23b2o27b3o$2o
72b2o$4b2o97b3o51bo$5b2o81bo14bo53bo$4bo82bo16bo52bo9bo$87b3o76b2o$
166bobo6$143b2o$143b2o4$149b2o$149b2o$144b3o$146bo$145bo$90b2o$90bobo$
72b2o16bo$73b2o$72bo!

[Freywa]

The following oscillators may be interesting to synthesise, or where a synthesis is known it may be interesting to improve that synthesis:

• R2D2
• 35P12.1
• Snake pit
• The stator variant of diamond ring which Summers synthesised
• Dinner table
• Fire-spitting

[mniemiec]

Six Ls in 20 gliders: ...

Very nice (and that includes the other syntheses as well)! Improved cleanup reduces this to 18:

Code: Select all

x = 126, y = 61, rule = B3/S23
69bo$67bobo$68boo7$92bo11bo$91bo10boo$74bo16b3o9boo$75boo22bo$74boo22b
o$98b3o8$21bo$21bobo$21boo$19bo$17bobo29boo38boo$18boo28bobbo36bobbo
27booboo$49boo38boobboo7b3o15bobobo$93bobo6bo17bobobo$60bo9bo22bo9bo
14boo4boo$61bo6bobo48bobobo$33boo24b3o7boobboo44bobobo$4boo26bobbo36bo
bbo44booboo$4bobo26boo38boo$4bo$boo$obo$bbo8$63b3o$65bo22boo$64bo22boo
$59boo9b3o16bo$60boo10bo$59bo11bo7$94boo$94bobo$94bo!

35P12.1 was discovered (I think) by Matthias on 2015-12-16, and synthesized from 21 gliders by Extrementhusiast;
he also synthesized Firespitting from 61 gliders in the same post:
http://conwaylife.com/forums/viewtopic. ... 954#p25954

Jason Summers' 28-glider synthesis of Diamond Ring variant (date unknown, but before 2013-06-02):

Code: Select all

x = 143, y = 43, rule = B3/S23
87bobo$88boo3bo$9bo78bo5boo$10bo82boo10bo$8b3o94bobo$105boo$$58bo39bo$
34bo23bo39bo$32boo24bo39bo$33boo$12bo63bo37bo$12boo13bo32boo15boo17bo
3boo11bo$11bobo13bobo30boo14boo19booboo11b3o$27boo67boo$$25bo108booboo
$26boo37boo10bobo25boo28bobobo$13b3o9boo38boo11boo25boo25bobbo3bo$15bo
62bo53b3obbobboo$14bo54b3o16bobo14bo3b3o25bo4bo$89boo13boo26boobooboob
oo$20bo22b3o37b3o3bo14bobo25bo4bo$19bo96bo16boobbobb3o$8boo9b3o26boo
38boo25boo18bo3bobbo$7boo39boo38boo25bobo17bobobo$9bo126booboo$$6boo
89boo$5bobo13bobo29boo24b3o11booboo19boo$7bo13boo30boo26bo11boo3bo17b
oo$22bo57bo37bo$oo$boo53bo39bo$o55bo39bo$56bo39bo$$88boo$24b3o60bobo$
24bo64bo10boo$25bo73boo5bo$101bo3boo$105bobo!

[Freywa]

As usual the git folder has been updated with synthesis files suitable for LifeWiki - do check them out. As of this post, running C2_1 on apgsearch in hopes of finding a snake pit soup.

[Goldtiger997]

You gave me exactly the right idea. Pentoad 2 in 16 gliders:

Code: Select all

x = 68, y = 37, rule = B3/S23
28bo$21bobo2bobo$22b2o3b2o$22bo10bobo$33b2o$9bo24bo$7b2o$8b2o$25b2o5bo
$6bo17bobo6bo2b3o$7bo18bo4b3o23b2o$5b3o49b2o2$39bo18bo$7b2o29bobo16b2o
$7bobo28bobo15bobo$7bo31bo16b2o$63bo$60b4o$60bo$2bo27bo35b2o$obo26bobo
33bobo$b2o26bobo33b2o$30bo34bo2$2b3o60b2o$2bo33b3o4bo21b2o$3bo27b3o2bo
6bobo$37bo5b2o$2o$b2o$o34bo$35b2o$34bobo10bo$41b2o3b2o$41bobo2bobo$41b
o!

Here's a reduction to 12 gliders, found with assistance from chris_c's popseq script:

Code: Select all

x = 62, y = 39, rule = B3/S23
61bo$40bo18b2o$41b2o17b2o$40b2o2$54bo$44bo8bo$45b2o6b3o$44b2o5$15bo$
14bo39bo$14b3o36bobo$53bobo$17b3o34bo$17bo$bo16bo$2bo$3o42bo$44bobo$3b
3o38bobo$5bo39bo$4bo5$54b2o$44b3o6b2o$46bo8bo$45bo2$58b2o$38b2o17b2o$
39b2o18bo$38bo!

[Freywa]

Obvious synthesis of 6 bits:

Code: Select all

x = 89, y = 45, rule = B3/S23
37bobo$37b2o$38bo2$67b3o$67bobo$33b2o5bo26b3o$34b2o3bo27b3o$33bo5b3o
25b3o$67b3o$67bobo$67b3o4$67bo$68b2o$16bobo48b2o$17b2o$17bo23bo$39bobo
8b8o23b8o$23b2o15b2o8bob4obo23bob4obo$22b2o26b8o23b8o$16b2o6bo$16bobo$
16bo4$42b3o28b3o$43bo30bo$43bo30bo$42b3o28b3o$obo$b2o39b3o28b3o$bo8b2o
30b3o28b3o$10bobo$10bo31b3o28b3o$43bo30bo$43bo30bo$42b3o28b3o2$8b2o$7b
2o$9bo!